Poll for differences between M9 vs M240 images- Red Dot Camera

Discussion in 'Leica News and Rumors' started by Brian, Feb 25, 2015.

  1. Brian

    Brian Super Moderator Subscribing Member

    Apr 3, 2013
    David Farkas has posted this poll for M9 vs M240, CMOS vs CCD: a can you pick the M9.

    http://www.reddotforum.com/content/2015/02/the-great-debate-ccd-vs-cmos-part-1/

    My guesses:

    M9:
    Lombard-Street-2
    Walkway-with-Puple-Flowers-2
    Apartments-1
    Gate-2
    Modern-Building-2
    Skyscraper-2
    Slippers-2
    Red-Cards-2
    Scarves-1
    Church-Windows-2
    Pier-39-1
    Steps-2
    Magenta-Tulips-1
    Sailboats-in-front-of-Alcatraz-1
    Bay-Boat-Tour-2
    Fishing-Boats-Bows-2
    Life-Preserver-1
    Fishing-Dock-1
    Streetcar-1
     
  2. ajramirez

    ajramirez Leica Place Veteran

    267
    Mar 30, 2013
    Caguas, Puerto Rico
    Antonio
    I participated as well. I am curious to see the actual results. I think that a bit of a push of the clarity slider (or some wide radius unsharp masking) would easily bring the results close enough to not make a difference. For me, the differences are small enough not to matter.

    Cheers,

    Antonio
     
  3. Christilou

    Christilou Leica Place Veteran

    358
    Apr 5, 2013
    Interesting, I thought it would be obvious but I found it very difficult in most cases. There were a few that screamed M9 but on the whole, not much difference.
     
  4. Brian

    Brian Super Moderator Subscribing Member

    Apr 3, 2013
    "Everything can be corrected in Post"... With the M9, I do straight exports to JPEG using Lightroom, very little post processing. With the M Monochrom- I use filters to get the look I want, no post processing- just straight export to JPEG.

    With the M8- I'm hooked in M8Raw, and just like pushing the camera beyond what the firmware could ever do.
     
  5. Brian

    Brian Super Moderator Subscribing Member

    Apr 3, 2013
    I had more fun with Part 1, where an image was posted from each camera. The M9 images "overall" took more space to store in the JPEG compared with the M240. The images were all hand-held, and taken at different times- so some "eyeballing" was required to see how the images changed. For whatever reason, the M9 images required more frequency coefficients to store, ie means more frequency content.

    Part two- only one image posted, here it goes:

    Part 2:

    1) M240 Cigar Store
    2) M240 Washington Square
    3) M240 Julie
    4) M240 Tulips
    5) M9- Two Guys.
    6) M9- Sign
    7) M240 Red Guage
    8) M240 Sub Dial
    9) M240 Sub-Engine Guage
    10) M240 Battle-Telephone
    11) M240 Conning Tower
    12) M240 Leaving Museum
    13) M240 Pier Sunset
    14) M240 Square at Night
    15) M240 Sunset on Beach
    16) M240 Sophia with hat
    17) M240 Angled Street
    18) M240 City-View
    19) M240 Coit Tower
    20) M240 Coit Tower UP
    21) M240 Red Car
    22) M240 Modern House
    23) M240 Loading Bay
    24) M240 Parking
    25) M240 Hallway
    26) M240 Fishmonger
    27) M240 Drydock
    28) M240 Embarcadero
    29) M240 Roll Your Own
    30) M240 Red Door
    31) M240 Shadow Tree

    There is one minor difference in image size when an M9 image and an M240 image is scaled to 1800 pixels on the long side. M9 comes to 1198, M240 to 1202. Easily changed for a test, I would have cropped them to all the same size. David could have swapped the numbers to throw such a measure off.
     
  6. Christilou

    Christilou Leica Place Veteran

    358
    Apr 5, 2013
    Second lot were even harder. I was looking more at the rendition in the blues/greens but to my eye most of them were M240 output. I probably put about 6 or 7 as M9. I have no idea how to see the exif of each in any case.
     
    • Like Like x 1
  7. Brian

    Brian Super Moderator Subscribing Member

    Apr 3, 2013
    The EXIF information was removed from all of the pictures. The first set- much more interesting information about the differences in the two cameras. The Leica Forum had such a test a while ago with an M9 image and M240 image that were taken on a tripod, images well-registered. The M9 image took more disk space to store the JPEG. Both images were stored with maximum details, ie Lossless. The M9 image had more frequency information in it. I'm not sure why, it could be an IR bleed problem: the M9 IR filter is stronger than the M240. An interesting test would be for some M240 owner to mount the camera on a tripod and take a series of tests with/without an IR cut filter over the lens. If the JPEG files require more space to store for the images used with the filter, then I would think IR blur is a possibility.
     
  8. Brian

    Brian Super Moderator Subscribing Member

    Apr 3, 2013
    I got 1 wrong in Group 1. Looks like JPEG Frequency content means something.

    I got all of Group 2 correct, but i went solely by the size ratio.
     
  9. Brian

    Brian Super Moderator Subscribing Member

    Apr 3, 2013
    http://www.reddotforum.com/content/2015/03/the-great-debate-ccd-vs-cmos-part-3/

    So I got one wrong in the first group, none in the second.

    The first group was more interesting as pictures from each camera were posted. The M9 images took more space on disk to store the JPEG, which means the images required more frequencies to reproduce. I think the difference is due to IR bleed in the M240, but that remains to be tested. the M9 IR filter is very efficient. Some M240 users keep an IR cut filter on the lens for best color reproduction. I use one with the M9 for my fastest lenses.
     
  10. Amin Sabet

    Amin Sabet Administrator

    214
    Feb 18, 2015
    +1
     
  11. Brian

    Brian Super Moderator Subscribing Member

    Apr 3, 2013
    The M9 images were done using Lossy compression for DNG. The M240 images were lossless.

    The M9 compression routine throws away spatial resolution, microcontrast, and increases peak-to-peak noise. It creates harsh edges where there were none im the uncompressed image. That threw my algorithm: all it did was detect the lossy compression used.

    As far as differences in CMOS and CCD: there are differences as they are very different devices. To really compare the two cameras, the uncompressed DNG files need to be provided, with lens detection set to "off". Then, you could compare the two with regard to sensor performance.

    I believe the future for M cameras is in Back-Side illuminated CMOS. Solves the geometry problem of the short lens register.

    As far as "turn-over rate" of generations of cameras, I am having a blast with the M8. After 8 years, it's potential is being realized.